传感网环境下事件驱动的林火动态观测方法研究

发布时间：2018-05-07 04:06

本文选题：森林火灾 + 观测事件　；参考：《武汉大学》2013年博士论文

【摘要】：火灾是最经常、最普遍地威胁公众安全和社会发展的主要灾害之一。全世界每年发生大量的森林火灾,给生态环境造成严重破坏,给人民生命财产造成巨大损失。如何对森林进行全面观测,及时获取环境信息,对林火的发生进行预测预报,以求尽可能地降低火灾造成的损失,一直是人类关心的重要研究课题。传统的观测以人工和半自动化的手段为主,观测过程比较繁琐,信息处理的效率比较低,不利于火险信息的及时发现。随着对地观测技术的发展,越来越多的空、天、地传感器加入了环境观测的行列,使人类能够更快地获取森林环境信息。而自上世纪末开始的计算机、通讯、微电子等领域的快速发展和融合,催生了传感网技术。这种技术整合了环境观测所需的感知、存储、通讯、网络、计算等等资源,可以大大提高空间信息的发现、获取和处理的自动化水平,从而提高环境观测的效率。在过去的十多年中,传感网技术获得了迅速地发展,在许多领域得到了研究和应用。但是,如何将传感网技术应用于森林环境观测,提高林火观测的水平,仍然是个有待深入研究的课题。在地学领域,传感网技术的研究和应用大多以OGC发展的系列规范为基础。OGC为了促进地理信息的开放和互操作,发展了多个系列的服务规范,其中包括地学信息服务(OGC Web Service)系列规范和传感网服务(Sensor Web Service)系列规范。这些服务规范均针对面向服务的架构(Service Oriented Architecture, SOA)设计。SOA架构可以很好地整合现有的资源,使它们协同工作,但是本身最适合的的模式是请求-响应,这种模式缺乏反应性。而环境观测是个感知的过程,除了被动的获取环境信息以外,主动地对环境变化作出反应也很重要。为此,OGC试图为系列服务规范引入事件驱动的方法,以弥补这方面的不足。经过十多年的努力,现有的研究成果已经能够支撑一些基于事件驱动的观测应用。目前,国内外与之相关的应用研究还比较少。论文从林火动态化观测的需求出发,研究在传感网环境下使用事件驱动方法观测林火所需的一些关键技术。主要工作和研究成果如下： (1)针对事件驱动的林火观测方法的需求,回顾和总结了林火观测的基本原理和方法、OGC系列规范中与传感网应用有关的服务规范的研究和发展、事件驱动技术的研究和发展、以及传感网中对事件驱动技术的研究和发展,指出了在传感网环境下实现事件驱动应用的基本思路。 (2)林火观测事件的分层、分类和建模方法。它们是事件驱动技术应用的设计基础。事件存在多种层次,但是人们在日常生活中谈论的事件并无明确的层次区分。而事件处理中这种区分是必要的,因为不同层次的事件处理方法可能会不一样。分析了观测与环境之间的层次关系,给出了观测事件的三种抽象层次。以一种林火预测模型为基础,研究了林火观测过程中的事件,并对它们进行了分类。在此基础上,建立了观测事件的概念模型。 (3)林火观测事件的处理方法。事件处理是事件驱动的关键。论文主要研究了观测事件处理中的几项关键技术,包括观测事件的组织、观测事件的过滤、复杂观测事件的处理和观测事件处理流程的建模。分析了传感网系列规范中对这些关键技术的支持。然后以一种林火预测模型为基础,研究了前三种技术的实现方法。最后采用有限状态机对观测事件处理流程进行了建模。 (4)适用于林火观测的事件驱动架构。事件驱动架构是设计和实现事件驱动应用的基础,大致包含四个部分：事件产生、事件通道、事件处理和事件消费。论文针对林火观测的需要,分析了这四部分中必需的构件。然后在此基础上构建并详细分析了观测事件的处理流程。接着,以该流程和现有数据服务设施为基础,设计了观测事件驱动的概念框架。最后,以概念框架为基础,采用传感网相关技术,设计了一种适用于林火观测的事件驱动架构。 (5)观测事件和事件模式的编码。基于观测与测量(Observations and Measurements, OM)规范对观测信息进行编码和基于事件模式描述语言(Event Pattern Markup Language, EML)对事件模式进行编码是传感网技术的研究热点和远景目标。论文采用现有的OM规范对观测事件进行编码,采用EML规范对事件模式进行编码,并使二者配合完成了事件过滤。 (6)针对一种林火预测方案,建立了仿真的林火观测实验平台,验证了论文提出的模型和方法的可行性。实验根据事件驱动架构划分为四个阶段,每个阶段分别测试了一些关键技术。此外,通过实验平台中各模块协同工作的时序图说明了事件驱动架构的运行过程。
[Abstract]:Fire is one of the most frequent and most common hazards to public safety and social development. A large number of forest fires occur every year in the world, causing serious damage to the ecological environment and causing huge losses to the people's lives and property. In order to reduce the loss caused by fire as much as possible, it has always been an important research subject of human concern. The traditional observation is mainly based on artificial and semi automated means, the process of observation is more complicated and the efficiency of information processing is low, which is not conducive to the timely discovery of fire insurance information. With the development of earth observation technology, more and more empty, sky, ground Sensors have joined the ranks of environmental observations so that human beings can get the information of the forest environment faster. The rapid development and integration of computers, communications, microelectronics and other fields, which have been developed from the end of the last century, have produced the sensing network technology. This technology integrates the sense of the environment, the storage, the communication, the network, the computing and so on. In the past more than 10 years, sensing network technology has developed rapidly in the past more than 10 years, and has been studied and applied in many fields. However, how to apply sensor network technology to forest environment observation and improve the level of forest fire observation is still to be improved. It is a subject to be studied in depth.
In the field of Geoscience, the research and application of sensor network technology is mostly based on the series of specifications developed by OGC. In order to promote the opening and interoperation of geographic information,.OGC has developed a number of series of service specifications, including the OGC Web Service series specification and the Sensor Web Service (Sensor Web Service) series specification. Service Oriented Architecture (SOA) design.SOA architecture can integrate existing resources well and work together, but the most suitable mode is request response, which is lack of reactivity. Environmental observation is a perceptual process, in addition to passive access to environmental information. In addition, it is important to actively respond to environmental changes. To this end, OGC attempts to introduce event driven approaches to a series of service specifications to make up for this problem. After more than 10 years of effort, the existing research results have been able to support event driven observation applications. From the demand of dynamic observation of forest fire, the paper studies the key technologies needed to observe forest fire using event driven method in sensing network environment. The main work and research results are as follows:
(1) in view of the demand of event driven forest fire observation method, the basic principles and methods of forest fire observation are reviewed and summarized, the research and development of service specification related to sensor network application in OGC series specifications, the research and development of event driven technology, and the research and development of event driven technology in the sensor network are also pointed out in the sensor network. The basic idea of realizing event driven application in the environment.
(2) the stratification, classification and modeling methods of forest fire observation events. They are the basis of the design of event driven technology applications. There are various levels of events, but there is no clear level of distinction between events in daily life. And this distinction is necessary in event processing, because different levels of event handling may be different. The hierarchical relationship between observation and environment is analyzed. Three kinds of abstract levels of observation events are given. Based on a forest fire prediction model, the events in the process of forest fire observation are studied and classified. On the basis of this, the conceptual model of observation events is established.
(3) the processing method of forest fire observation events. Event processing is the key to event driven. This paper mainly studies several key technologies in the observation event processing, including the organization of observation events, the filtering of observation events, the processing of complex observation events and the modeling of the process of observation event processing. The key technology is supported. Then, based on a forest fire prediction model, the realization methods of the first three technologies are studied. Finally, the finite state machine is used to model the process of observation event processing.
(4) event driven architecture for forest fire observation. Event driven architecture is the basis for the design and implementation of event driven applications. It contains roughly four parts: event generation, event channel, event handling and event consumption. The paper analyzes the necessary components in the four parts for the needs of forest fire observation. The process of observation event processing is analyzed. Then, based on the process and existing data service facilities, the conceptual framework of observation event driven is designed. Finally, based on the conceptual framework, a sensor network related technology is used to design an event driven architecture suitable for forest fire observation.
(5) coding of observational events and event patterns. Coding based on observation and measurement (Observations and Measurements, OM) specification and event mode description language (Event Pattern Markup Language, EML) is the research hotspot and Prospect of sensor network technology. The existing OM is used in this paper. Code is used to encode the observation events, and the EML code is used to encode the event patterns, and the two parties cooperate to complete the event filtering.
(6) in view of a forest fire prediction scheme, a simulated forest fire observation experiment platform is set up to verify the feasibility of the model and method proposed in this paper. The experiment is divided into four stages according to the event driven architecture, and some key technologies are tested at each stage. Furthermore, the time series diagram of the coordination work of each module in the testing platform is illustrated. The running process of the event driven architecture.

【学位授予单位】：武汉大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：S762;P208

【参考文献】